Update README.md

README.md

@@ -10,13 +10,15 @@ tags:
 - on-device
 - m4-max
 - text-to-speech
+base_model:
+- vibevoice/VibeVoice-7B
 ---
 
 # VibeVoice 7B — multi-speaker TTS (Core AI)
 
-Highest-quality multi-speaker TTS, 7B Qwen2 backbone. Bundles its own σ-VAE codec (shared with 1.5b) so the repo is self-contained. cfg 2.0 fused sampler avoids long-form static buildup.
+High-quality multi-speaker TTS, 7B Qwen2 backbone. cfg 2.0 fused sampler avoids long-form static buildup.
 
-**On-device performance (M4 Max, Core AI):** 2.37× RTF, WER 0%.
+**On-device performance (M4 Max, Core AI):** 2.37× RTF.
 
 > ⚠️ **Beta artifacts.** These `.aimodel` bundles are compiled for macOS 27 / Xcode 27 beta (Core AI). They may need re-export on the GA toolchain. The original weights are Microsoft VibeVoice (see upstream for the model license).
 
@@ -82,4 +84,4 @@ Resolve assets by role via `manifest.json` (`default` = recommended variant):
   "decode_compute": "gpu",
   "sem_compute": "gpu"
 }
-```
+```